Application of machine learning in predicting the risk of postpartum depression: A systematic review

Development of an explainable machine learning model for predicting depression in adolescent girls with non-suicidal self-injury: A cross-sectional multicenter study

Non-suicidal self-injury (NSSI) in adolescent girls is a critical predictor of subsequent depression and suicide risk, yet current tools lack both accuracy and clinical interpretability. We developed the first explainable machine learning model integrating multicenter psychosocial data to predict depression among Chinese adolescent girls with NSSI, addressing the critical need for culturally tailored risk stratification tools. In this cross - sectional observational study, our model was developed using data from 14 hospitals. We used five categories of data as predictors, including individual, family, school, psychosocial, and behavioral and lifestyle factors. We compared seven machine learning models and selected the best one to develop final model and the Shapley Additive exPlanations (SHAP) method were used to explain model prediction. The Random Forest (RF) model was compared against six other machine learning algorithms. We assessed the discrimination using the area under receiver operating characteristic (AUROC) with 95 % CIs. Using the development dataset (n = 1163) and predictive model building process, a simplified model containing only the top 20 features had similar predictive performance to the full model, the RF model outperformed six algorithms (AUROC = 0.964 [0.945-0.975]), demonstrating superior discriminative power and robustness. The top ten risk predictors were Borderline personality, Rumination, Perceived stress, Hopelessness, Self-esteem, Sleep quality, Loneliness, Resilience, Parental care, and Problem-focused coping. We developed a three-tiered, color-coded web-based clinical tool to operationalize predictions, enabling real-time risk stratification and personalized interventions. Our study bridges machine learning and clinical interpretability to advance precision mental health interventions for vulnerable adolescent populations.

Artificial Intelligence for Women and Child Healthcare: Is AI Able to Change the Beginning of a New Story? A Perspective

Background and Aims

Maternal and neonatal mortality remain critical global health challenges, particularly in low-resource settings where preventable deaths occur due to inadequate access to timely care. This article explores the potential of Artificial Intelligence (AI) to enhance maternal and child healthcare by improving early risk identification, diagnosis, treatment recommendations, and postpartum monitoring.

Methods

It explores the use of AI in identifying pregnancy-related risks, recommending treatments, predicting adverse outcomes, and monitoring postpartum and neonatal care. Various AI models, including supervised machine learning, Large Language Models (LLMs), and Small/Medium Language Models (SLMs/MLMs), are discussed in terms of their feasibility into resource-limited healthcare systems.

Results

AI has demonstrated significant potential in identifying pregnancy-related risks, recommending treatments, predicting adverse outcomes, and supporting postpartum and neonatal care. While AI-driven solutions can optimize healthcare decision-making and resource allocation, challenges such as data availability, integration into clinical workflows, and ethical considerations must be addressed for widespread adoption.

Conclusion

AI offers promising solutions to reduce maternal and neonatal mortality by enhancing risk detection and clinical decision-making. However, its real-world implementation requires overcoming barriers related to data quality, infrastructure, and equitable deployment. Future efforts should focus on data standardization, AI model optimization for resource-limited settings, and ethical considerations in clinical integration.

Text-Based Depression Prediction on Social Media Using Machine Learning: Systematic Review and Meta-Analysis

BACKGROUND

Depression affects more than 350 million people globally. Traditional diagnostic methods have limitations. Analyzing textual data from social media provides new insights into predicting depression using machine learning. However, there is a lack of comprehensive reviews in this area, which necessitates further research.

OBJECTIVE

This review aims to assess the effectiveness of user-generated social media texts in predicting depression and evaluate the influence of demographic, language, social media activity, and temporal features on predicting depression on social media texts through machine learning.

METHODS

We searched studies from 11 databases (CINHAL [through EBSCOhost], PubMed, Scopus, Ovid MEDLINE, Embase, PubPsych, Cochrane Library, Web of Science, ProQuest, IEEE Explore, and ACM digital library) from January 2008 to August 2023. We included studies that used social media texts, machine learning, and reported area under the curve, Pearson r, and specificity and sensitivity (or data used for their calculation) to predict depression. Protocol papers and studies not written in English were excluded. We extracted study characteristics, population characteristics, outcome measures, and prediction factors from each study. A random effects model was used to extract the effect sizes with 95% CIs. Study heterogeneity was evaluated using forest plots and P values in the Cochran Q test. Moderator analysis was performed to identify the sources of heterogeneity.

RESULTS

A total of 36 studies were included. We observed a significant overall correlation between social media texts and depression, with a large effect size (r=0.630, 95% CI 0.565-0.686). We noted the same correlation and large effect size for demographic (largest effect size; r=0.642, 95% CI 0.489-0.757), social media activity (r=0.552, 95% CI 0.418-0.663), language (r=0.545, 95% CI 0.441-0.649), and temporal features (r=0.531, 95% CI 0.320-0.693). The social media platform type (public or private; P<.001), machine learning approach (shallow or deep; P=.048), and use of outcome measures (yes or no; P<.001) were significant moderators. Sensitivity analysis revealed no change in the results, indicating result stability. The Begg-Mazumdar rank correlation (Kendall τb=0.22063; P=.058) and the Egger test (2-tailed t34=1.28696; P=.207) confirmed the absence of publication bias.

CONCLUSIONS

Social media textual content can be a useful tool for predicting depression. Demographics, language, social media activity, and temporal features should be considered to maximize the accuracy of depression prediction models. Additionally, the effects of social media platform type, machine learning approach, and use of outcome measures in depression prediction models need attention. Analyzing social media texts for depression prediction is challenging, and findings may not apply to a broader population. Nevertheless, our findings offer valuable insights for future research.

TRIAL REGISTRATION

PROSPERO CRD42023427707; https://www.crd.york.ac.uk/PROSPERO/view/CRD42023427707.

Targeted Research and Treatment Implications in Women With Depression

Women with a history of traumatic experience, particularly adversity encountered during childhood, have an increased risk of developing depression. The authors review the biological mechanisms associating trauma with depression, including the role of the hypothalamic-pituitary-adrenal axis. Additionally, the psychosocial and cultural considerations associating traumatic experience with depression are discussed, and current gaps in knowledge about biological mechanisms, psychosocial factors, and cultural aspects relating trauma to depression that remain to be addressed are described. Women with a history of trauma are also at increased risk for engaging in suicidal behaviors, including suicidal ideation and attempts. Increased suicidality in women with a history of trauma has been observed in various populations, including among victims of intimate partner violence, female veterans, refugees, and individuals who identify as lesbian, gay, bisexual, transgender, queer or questioning, or other. Although associations between trauma and suicidality have been well documented, limited research has examined the impact of age or reproductive stage, an important area for future research. A wide range of biological, psychosocial, and cultural factors that can increase the risk for suicidality across the lifespan in women are described, and how they may be included when completing clinical assessments for women is highlighted. Machine learning, and its use in risk and outcome prediction of depression in women across reproductive stages toward individualized psychiatric services, is introduced, with future directions reviewed.

DMN network and neurocognitive changes associated with dissociative symptoms in major depressive disorder: a research protocol

Introduction

Depression is a heterogeneous disorder with diverse clinical presentations and etiological underpinnings, necessitating the identification of distinct subtypes to enhance targeted interventions. Dissociative symptoms, commonly observed in major depressive disorder (MDD) and linked to early life trauma, may represent a unique clinical dimension associated with specific neurocognitive deficits. Although emerging research has begun to explore the role of dissociation in depression, most studies have provided only descriptive analyses, leaving the mechanistic interplay between these phenomena underexplored. The primary objective of this study is to determine whether MDD patients with prominent dissociative symptoms differ from those without such symptoms in clinical presentation, neurocognitive performance, and markers of functional connectivity. This investigation will be the first to integrate comprehensive clinical evaluations, advanced neurocognitive testing, and high-resolution brain imaging to delineate the contribution of dissociative symptoms in MDD.

Methods

We will recruit fifty participants for each of three groups: (1) depressive patients with dissociative symptoms, (2) depressive patients without dissociative symptoms, and (3) healthy controls. Diagnostic assessments will be performed using the Structured Clinical Interview for DSM-5 (SCID) alongside standardized scales for depression severity, dissociation, and childhood trauma. Neurocognitive performance will be evaluated through a battery of tests assessing memory, attention, executive function, and processing speed. Structural and functional magnetic resonance imaging (MRI) will be conducted on a 3 Tesla scanner, focusing on the connectivity of the Default Mode Network with key regions such as the orbitofrontal cortex, insula, and posterior cingulate cortex. Data analyses will employ SPM-12 and Matlab-based CONN and PRONTO tools, with multiclass Gaussian process classification applied to differentiate the three groups based on clinical, cognitive, and imaging data.

Discussion

The results of this study will introduce a novel perspective on understanding the connection between major depressive disorder and dissociation. It could also aid in pinpointing a distinct form of depression associated with dissociative symptoms and early childhood stressors.

Conclusion

Future research, aiming to forecast the response to biological and psychological interventions for depression, anticipates this subtype and provides insights.

Application of Predictive Analytics in Pregnancy, Birth, and Postpartum Nursing Care

ABSTRACT

Predictive analytics has emerged as a promising approach for improving reproductive health care and patient outcomes. During pregnancy and birth, the ability to accurately predict risks and complications could enable earlier interventions and reduce adverse events. However, there are challenges and ethical considerations for implementing predictive models in perinatal care settings. We introduce major concepts in predictive analytics and describe application of predictive modeling to perinatal care topics such as fertility, preeclampsia, labor onset, vaginal birth after cesarean, uterine rupture, induction outcomes, postpartum hemorrhage, and postpartum mood disorders. Although some predictive models have achieved adequate accuracy (AUC 0.7-0.9), most require additional external validation across diverse populations and practice settings. Bias, particularly racial bias, remains a key limitation of current models. Nurses and advanced practice nurses, including nurse practitioners certified registered nurse anesthetists, and nurse-midwives, play a vital role in ensuring high-quality data collection and communicating predictive model outputs to clinicians and users of the health care system. Addressing the ethical challenges and limitations of predictive analytics is imperative to equitably translate these tools to support patient-centered perinatal care.

Decision tree learning for predicting chronic postpartum depression in the Japan Environment and Children's Study

BACKGROUND

Many studies have used machine learning techniques to construct predictive models of postpartum depression, but few such models are simple enough to use in community maternal health settings with pen and paper. Here, we use a decision tree to construct a prediction model for chronic postpartum depression.

METHODS

Participants were 84,091 mothers. Chronic postpartum depression was identified as an Edinburgh Postnatal Depression Scale score of ≥9 at both 1 and 6 months postpartum. The training dataset included 84 diverse variables measured during pregnancy, including health status and biomarkers. In learning, the branching depth was constrained to 3, the number of branches per branch to 4, and the minimum number of n in a branch was 100. The training to validation data ratio was set to 7:3.

RESULTS

A decision tree with 35 branches and an area under the receiver operating characteristic of 0.84 was created. Ten of 84 variables were extracted, and the most effective in classification was "feeling worthless." At training (n = 58,635), the most and least prevalent branches were 73.2 % and 0.84 % (mean = 6.29 %), respectively; at validation (n = 25,456), they were 60.4 % and 0.72 % (mean = 6.52 %), respectively.

LIMITATIONS

Chronic postpartum depression was identified using self-administered questionnaires.

CONCLUSIONS

This study created a simple and relatively high-performing prediction model. Because the model can be easily understood and used without expertise in machine learning, it is expected to be useful in maternal health settings, including grassroots community health.

Consumer opinion on the use of machine learning in healthcare settings: A qualitative systematic review

Introduction

Given the increasing number of artificial intelligence and machine learning (AI/ML) tools in healthcare, we aimed to gain an understanding of consumer perspectives on the use of AI/ML tools for healthcare diagnostics.

Methods

We conducted a qualitative systematic review, following established standardized methods, of the existing literature indexed in the following databases up to 4 April 2022: OVID MEDLINE, OVID EMBASE, Scopus and Web of Science.

Results

Fourteen studies were identified as appropriate for inclusion in the meta-synthesis and systematic review. Most studies (n = 12) were conducted in high-income countries, with data extracted from both mixed methods (42.9%) and qualitative (57.1%) studies. The meta-synthesis identified four overarching themes across the included studies: (1) Trust, fear, and uncertainty; (2) Data privacy and ML governance; (3) Impact on healthcare delivery and access; and (4) Consumers want to be engaged.

Conclusion

The current evidence demonstrates consumers' understandings of AI/ML for medical diagnosis are complex. Consumers express a complex combination of both hesitancy and support towards AI/ML in healthcare diagnosis. Importantly, their views of the use of AI/ML in medical diagnosis are influenced by the perceived trustworthiness of their healthcare providers who use these AI/ML tools. Consumers recognize the potential for AI/ML tools to improve diagnostic accuracy, efficiency and access, and express a strong interest to be engaged in the development and implementation process of AI/ML into routine healthcare.

Predicting first time depression onset in pregnancy: applying machine learning methods to patient-reported data

PURPOSE

To develop a machine learning algorithm, using patient-reported data from early pregnancy, to predict later onset of first time moderate-to-severe depression.

METHODS

A sample of 944 U.S. patient participants from a larger longitudinal observational cohortused a prenatal support mobile app from September 2019 to April 2022. Participants self-reported clinical and social risk factors during first trimester initiation of app use and completed voluntary depression screenings in each trimester. Several machine learning algorithms were applied to self-reported data, including a novel algorithm for causal discovery. Training and test datasets were built from a randomized 80/20 data split. Models were evaluated on their predictive accuracy and their simplicity (i.e., fewest variables required for prediction).

RESULTS

Among participants, 78% identified as white with an average age of 30 [IQR 26-34]; 61% had income ≥ $50,000; 70% had a college degree or higher; and 49% were nulliparous. All models accurately predicted first time moderate-severe depression using first trimester baseline data (AUC 0.74-0.89, sensitivity 0.35-0.81, specificity 0.78-0.95). Several predictors were common across models, including anxiety history, partnered status, psychosocial factors, and pregnancy-specific stressors. The optimal model used only 14 (26%) of the possible variables and had excellent accuracy (AUC = 0.89, sensitivity = 0.81, specificity = 0.83). When food insecurity reports were included among a subset of participants, demographics, including race and income, dropped out and the model became more accurate (AUC = 0.93) and simpler (9 variables).

CONCLUSION

A relatively small amount of self-report data produced a highly predictive model of first time depression among pregnant individuals.

Predicting the Risk of Loneliness in Children and Adolescents: A Machine Learning Study

BACKGROUND

Loneliness is increasingly emerging as a significant public health problem in children and adolescents. Predicting loneliness and finding its risk factors in children and adolescents is lacking and necessary, and would greatly help determine intervention actions.

OBJECTIVE

This study aimed to find appropriate machine learning techniques to predict loneliness and its associated risk factors among schoolchildren.

METHODS

The data were collected from an ongoing prospective puberty cohort that was established in Chongqing, Southwest China. This study used 822 subjects (46.84% boys, age range: 11-16) followed in 2019. Five models, (a) random forest, (b) extreme gradient boosting (XGBoost), (c) logistic regression, (d) neural network, and (e) support vector machine were applied to predict loneliness. A total of 39 indicators were collected and 28 predictors were finally included for prediction after data pre-processing, including demographic, parental relationship, mental health, pubertal development, behaviors, and environmental factors. Model performance was determined by accuracy and AUC. Additionally, random forest and XGBoost were applied to identify the important factors. The XGBoost algorithm with SHAP was also used to interpret the results of our ML model.

RESULTS

All machine learning performed with favorable accuracy. Compared to random forest (AUC: 0.87 (95%CI: 0.80, 0.93)), logistic regression (AUC: 0.80 (95%CI: 0.70, 0.89)), neural network (AUC: 0.80 (95%CI: 0.71, 0.89)), and support vector machine (AUC: 0.79 (95%CI: 0.79, 0.89)), XGBoost algorithm had the highest AUC values 0.87 (95%CI: 0.80, 0.93) in the test set, although the difference was not significant between models. Peer communication, index of general affect, peer alienation, and internet addiction were the top four significant factors of loneliness in children and adolescents.

CONCLUSIONS

The results of this study suggest that machine learning has considerable potential to predict loneliness in children. This may be valuable for the early identification and intervention of loneliness.

Comparative analysis of machine learning versus traditional method for early detection of parental depression symptoms in the NICU

Introduction

Neonatal intensive care unit (NICU) admission is a stressful experience for parents. NICU parents are twice at risk of depression symptoms compared to the general birthing population. Parental mental health problems have harmful long-term effects on both parents and infants. Timely screening and treatment can reduce these negative consequences.

Objective

Our objective is to compare the performance of the traditional logistic regression with other machine learning (ML) models in identifying parents who are more likely to have depression symptoms to prioritize screening of at-risk parents. We used data obtained from parents of infants discharged from the NICU at Children's National Hospital (n = 300) from 2016 to 2017. This dataset includes a comprehensive list of demographic characteristics, depression and stress symptoms, social support, and parent/infant factors.

Study design

Our study design optimized eight ML algorithms - Logistic Regression, Support Vector Machine, Decision Tree, Random Forest, XGBoost, Naïve Bayes, K-Nearest Neighbor, and Artificial Neural Network - to identify the main risk factors associated with parental depression. We compared models based on the area under the receiver operating characteristic curve (AUC), positive predicted value (PPV), sensitivity, and F-score.

Results

The results showed that all eight models achieved an AUC above 0.8, suggesting that the logistic regression-based model's performance is comparable to other common ML models.

Conclusion

Logistic regression is effective in identifying parents at risk of depression for targeted screening with a performance comparable to common ML-based models.

Prevalence and risk factors analysis of postpartum depression at early stage using hybrid deep learning model

Postpartum Depression Disorder (PPDD) is a prevalent mental health condition and results in severe depression and suicide attempts in the social community. Prompt actions are crucial in tackling PPDD, which requires a quick recognition and accurate analysis of the probability factors associated with this condition. This concern requires attention. The primary aim of our research is to investigate the feasibility of anticipating an individual's mental state by categorizing individuals with depression from those without depression using a dataset consisting of text along with audio recordings from patients diagnosed with PPDD. This research proposes a hybrid PPDD framework that combines Improved Bi-directional Long Short-Term Memory (IBi-LSTM) with Transfer Learning (TL) based on two Convolutional Neural Network (CNN) architectures, respectively CNN-text and CNN audio. In the proposed model, the CNN section efficiently utilizes TL to obtain crucial knowledge from text and audio characteristics, whereas the improved Bi-LSTM module combines written material and sound data to obtain intricate chronological interpersonal relationships. The proposed model incorporates an attention technique to augment the effectiveness of the Bi-LSTM scheme. An experimental analysis is conducted on the PPDD online textual and speech audio dataset collected from UCI. It includes textual features such as age, women's health tracks, medical histories, demographic information, daily life metrics, psychological evaluations, and 'speech records' of PPDD patients. Data pre-processing is applied to maintain the data integrity and achieve reliable model performance. The proposed model demonstrates a great performance in better precision, recall, accuracy, and F1-score over existing deep learning models, including VGG-16, Base-CNN, and CNN-LSTM. These metrics indicate the model's ability to differentiate among women at risk of PPDD vs. non-PPDD. In addition, the feature importance analysis demonstrates that specific risk factors substantially impact the prediction of PPDD. The findings of this research establish a basis for improved precision and promptness in assessing the risk of PPDD, which may ultimately result in earlier implementation of interventions and the establishment of support networks for women who are susceptible to PPDD.

Major Depressive Disorder Prediction Based on Sleep-Wake Disorders Symptoms in US Adolescents: A Machine Learning Approach from National Sleep Research Resource

Background

There is substantial evidence from previous studies that abnormalities in sleep parameters associated with depression are demonstrated in almost all stages of sleep architecture. Patients with symptoms of sleep-wake disorders have a much higher risk of developing major depressive disorders (MDD) compared to those without.

Objective

The aim of the present study is to establish and compare the performance of different machine learning models based on sleep-wake disorder symptoms data and to select the optimal model to interpret the importance of sleep-wake disorder symptoms to predict MDD occurrence in adolescents.

Methods

We derived data for this work from 2020 to 2021 Assessing Nocturnal Sleep/Wake Effects on Risk of Suicide Phase I Study from National Sleep Research Resource. Using demographic and sleep-wake disorder symptoms data as predictors and the occurrence of MDD measured base on the center for epidemiologic studies depression scale as an outcome, the following six machine learning predictive models were developed: eXtreme Gradient Boosting model (XGBoost), Light Gradient Boosting mode, AdaBoost, Gaussian Naïve Bayes, Complement Naïve Bayes, and multilayer perceptron. The models' performance was assessed using the AUC and other metrics, and the final model's predictor importance ranking was explained.

Results

XGBoost is the optimal predictive model in comprehensive performance with the AUC of 0.804 in the test set. All sleep-wake disorder symptoms were significantly positively correlated with the occurrence of adolescent MDD. The insomnia severity was the most important predictor compared with the other predictors in this study.

Conclusion

This machine learning predictive model based on sleep-wake disorder symptoms can help to raise the awareness of risk of symptoms between sleep-wake disorders and MDD in adolescents and improve primary care and prevention.

Optimal risk and diagnosis assessment strategies in perinatal depression: A machine learning approach from the life-ON study cohort

This study aimed to assess the concordance of various psychometric scales in detecting Perinatal Depression (PND) risk and diagnosis. A cohort of 432 women was assessed at 10-15th and 23-25th gestational weeks, 33-40 days and 180-195 days after delivery using the Edinburgh Postnatal Depression Scale (EPDS), Visual Analogue Scale (VAS), Hamilton Depression Rating Scale (HDRS), Montgomery-Åsberg Depression Rating Scale (MADRS), and Mini International Neuropsychiatric Interview (MINI). Spearman's rank correlation coefficient was used to assess agreement across instruments, and multivariable classification models were developed to predict the values of a binary scale using the other scales. Moderate agreement was shown between the EPDS and VAS and between the HDRS and MADRS throughout the perinatal period. However, agreement between the EPDS and HDRS decreased postpartum. A well-performing model for the estimation of current depression risk (EPDS > 9) was obtained with the VAS and MADRS, and a less robust one for the estimation of current major depressive episode (MDE) diagnosis (MINI) with the VAS and HDRS. When the EPDS is not feasible, the VAS may be used for rapid and comprehensive postpartum screening with reliability. However, a thorough structured interview or clinical examination remains necessary to diagnose a MDE.

A random forest algorithm-based prediction model for moderate to severe acute postoperative pain after orthopedic surgery under general anesthesia

BACKGROUND

Postoperative pain is one of the most common complications after surgery. In order to detect early and intervene in time for moderate to severe postoperative pain, it is necessary to identify risk factors and construct clinical prediction models. This study aimed to identify significant risk factors and establish a better-performing model to predict moderate to severe acute postoperative pain after orthopedic surgery under general anesthesia.

METHODS

Patients who underwent orthopedic surgery under general anesthesia were divided into patients with moderate to severe pain group (group P) and patients without moderate to severe pain group (group N) based on VAS scores. The features selected by Lasso regression were processed by the random forest and multivariate logistic regression models to predict pain outcomes. The classification performance of the two models was evaluated through the testing set. The area under the curves (AUC), the accuracy of the classifiers, and the classification error rate for both classifiers were calculated, the better-performing model was used to predict moderate to severe acute postoperative pain after orthopedic surgery under general anesthesia.

RESULTS

A total of 327 patients were enrolled in this study (228 in the training set and 99 in the testing set). The incidence of moderate to severe postoperative pain was 41.3%. The random forest model revealed a classification error rate of 25.2% and an AUC of 0.810 in the testing set. The multivariate logistic regression model revealed a classification error rate of 31.3% and an AUC of 0.764 in the testing set. The random forest model was chosen for predicting clinical outcomes in this study. The risk factors with the greatest and second contribution were immobilization and duration of surgery, respectively.

CONCLUSIONS

The random forest model can be used to predict moderate to severe acute postoperative pain after orthopedic surgery under general anesthesia, which is of potential clinical application value.

Predicting Patients Requiring Treatment for Depression in the Postpartum Period Using Common Electronic Medical Record Data Available Antepartum

Introduction

Depression requiring treatment in the postpartum period significantly impacts maternal and neonatal health. Although preventive management of depression in pregnancy has been shown to decrease the negative impacts, current methods for identifying at-risk patients are insufficient. Given the complexity of the diagnosis and interplay of clinical/demographic factors, we tested whether machine learning techniques can accurately identify at-risk patients in the postpartum period.

Methods

This is a retrospective cohort study of the NIH Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-Be, which enrolled 10,038 nulliparous people. The primary outcome was depression in the postpartum period. We constructed and optimized 4 machine learning models using distributed random forest modeling and 1 logistic regression model on the basis of the NIH Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-Be dataset. Model 1 utilized only readily obtainable sociodemographic data. Model 2 added maternal prepregnancy mental health data. Model 3 utilized recursive feature elimination to construct a parsimonious model. Model 4 further titrated the input data to simplify prepregnancy mental health variables. The logistic regression model used the same input data as Model 3 as a proof of concept.

Results

Of 8,454 births, 338 (4%) were complicated by depression in the postpartum period. Model 3 was the highest performing, showing the area under the receiver operating characteristics curve of 0.91 (±0.02). Models 1-3 identified the 9 variables most predictive of depression hierarchically, ranging from depression history (highest), history of mental health condition, recent psychiatric medication use, BMI, income, age, anxiety history, education, and preparedness for pregnancy (lowest). In Model 4, the area under the receiver operating characteristics curve remained at 0.79 (±0.05).

Conclusions

Postpartum depression can be predicted with high accuracy for individual patients using antepartum information commonly found in electronic medical records. In addition, baseline mental health status and sociodemographic factors have a larger role in the postpartum period than previously understood.

Challenges and Opportunities for Data Science in Women's Health

The intersection of women's health and data science is a field of research that has historically trailed other fields, but more recently it has gained momentum. This growth is being driven not only by new investigators who are moving into this area but also by the significant opportunities that have emerged in new methodologies, resources, and technologies in data science. Here, we describe some of the resources and methods being used by women's health researchers today to meet challenges in biomedical data science. We also describe the opportunities and limitations of applying these approaches to advance women's health outcomes and the future of the field, with emphasis on repurposing existing methodologies for women's health.

SVM-Based Model Combining Patients' Reported Outcomes and Lymphocyte Phenotypes of Depression in Systemic Lupus Erythematosus

BACKGROUND

The incidence of depression in patients with systemic lupus erythematosus (SLE) is high and leads to a lower quality of life than that in undepressed SLE patients and healthy individuals. The causes of SLE depression are still unclear.

METHODS

A total of 94 SLE patients were involved in this study. A series of questionnaires (Hospital Depression Scale, Social Support Rate Scale and so on) were applied. Flow cytometry was used to test the different stages and types of T cells and B cells in peripheral blood mononuclear cells. Univariate and multivariate analyses were conducted to explore the key contributors to depression in SLE. Support Vector Machine (SVM) learning was applied to form the prediction model.

RESULTS

Depressed SLE patients showed lower objective support, severer fatigue, worse sleep quality and higher percentages of ASC%PBMC, ASC%CD19+, MAIT, TEM%Th, TEMRA%Th, CD45RA+CD27-Th, TEMRA%CD8 than non-depressed patients. A learning-based SVM model combining objective and patient-reported variables showed that fatigue, objective support, ASC%CD19+, TEM%Th and TEMRA%CD8 were the main contributing factors to depression in SLE. With the SVM model, the weight of TEM%Th was 0.17, which is the highest among objective variables, and the weight of fatigue was 0.137, which was the highest among variables of patients' reported outcomes.

CONCLUSIONS

Both patient-reported factors and immunological factors could be involved in the occurrence and development of depression in SLE. Scientists can explore the mechanism of depression in SLE or other psychological diseases from the above perspective.